| With the rapid development of Global Navigation Satellite System(GNSS),GNSS technology has been widely used in automatic driving,modern agriculture and other fields.However,the high-precision positioning based on carrier differential positioning technology(RTK)depends on the regional reference station system,and has some problems such as limited service system capacity and need to pay for use,so it is difficult to achieve popular application.Single point positioning technology can realize precise positioning on a global scale without reference station,and can further improve the positioning accuracy through post-processing,which has important application prospects in agricultural survey,natural resources survey and other fields.Based on this,this paper evaluates the accuracy of GNSS data collected under different quality and observation environment based on pseudo-distance single point positioning technology and precise single point positioning technology respectively.Furthermore,the smooth technique of position domain and observation range domain is used to improve the positioning accuracy.The main work and conclusions of this paper are as follows:(1)Aiming at the problem that satellite signals are affected by multipath interference in complex environments and the positioning accuracy of low-cost receivers is poor at pseudo-distance single point,this paper adopts unfiltered and two kinds of position domain smoothing to solve the SPP positioning of the observation data collected by receivers of different quality in different environments,and verifies the accuracy improvement of smoothing algorithm.The experimental results show that the plane-positioning accuracy of low-cost receivers LC29 H and Sino K803 without filtering are 1.61 m,3.09 m and 1.88 m,respectively,in Trimble Alloy,a measured receiver under the environment of trees.The position smoothing algorithm is superior to the forward and backward filtering smoothing results.The planar accuracy of the three receivers using position smoothing is 1.32 m,2.56 m and 1.53 m,respectively,which is 18%,17% and 19% higher than that without filtering.In a relatively open environment,the plane precision of Huawei P40,LG69 T and Ublox-M8 T without filtering is 3.22 m,0.70 m and 1.59 m,respectively,and the plane precision of the three receivers solved by the position smoothing algorithm is 2.64 m,0.56 m and 1.25 m,respectively.Compared with the unfiltered solution,the results were improved by 28%,20% and 21%respectively.The relationship between the pseudo distance residual and the height Angle and the signal-to-noise ratio of LC29 H and Hudah Beidou HD in complex environment was analyzed,and the signal-to-noise ratio weight was adopted to solve the SPP positioning.The unfiltered plane accuracy of the two receivers were 3.58 m and 3.99 m,respectively,and the plane accuracy of the two receivers using position smoothing was 2.75 m and 3.09 m,respectively.Compared with the unfiltered solution,the improvement is 23% and 22%respectively.(2)In order to explore the difference of PPP positioning accuracy improvement based on observation range smoothing algorithm,this paper adopts the global least squares and reverse smoothing algorithm to solve the PPP positioning of the observation data collected by21 IGS stations in 4 days and receivers of different quality under different environments.The experimental results show that the PPP positioning results of most IGS stations using the global least squares algorithm are more accurate than the backward smooth solution.Among them,the plane accuracy of BIK0 on February 15 and CHAN station on February 16 using the global least squares algorithm is 0.54 cm and 0.65 cm.Compared with the reverse smoothing results,the results improved by 67% and 69%.The PPP positioning accuracy calculated by the global least squares of the measuring receiver Trimble and the low-cost receiver Sino in the environment of tree occlusion is better than the result of reverse smoothing,and the plane accuracy of the two receivers using the global least squares is 73%and 56% higher than that of reverse smoothing,respectively.(3)Further,different quality GNSS receivers are used to verify the difference in accuracy improvement of the location domain smoothing algorithm.The experimental results show that the planar accuracy of the measurement receiver Trimble and the low-cost receiver Sino in a unilateral occlusion environment using the global least square solution based on the observation range is 13.36 cm and 1.7 cm,respectively,and the planar accuracy of the two receivers using the position smoothing algorithm is 12.3 cm and 2.57 cm,respectively.The position smoothing algorithm can improve the positioning accuracy of Trimble.For low-cost receiver Sino,its anti multipath ability is poor,with high observation noise,making the position smoothing algorithm reduce its planar positioning accuracy instead.After using the position smoothing algorithm under tree occlusion,the positioning accuracy of Trimble and Sino planes is 2.23 cm and 4.52 cm,respectively,which is 23% and 7% higher than the dynamic PPP plane accuracy using the overall least square solution.According to the observation noise covariance matrix in the receiver’s quality adjusted position smoothing algorithm,the planar accuracy of Sino’s position smoothing solution after adjustment is better than that of the position smoothing solution before adjustment,and Trimble’s positioning results have no impact. |
